• Section properties use round corners for everything, including $J$ and $C_w$ (sharp corners are unconservative for torsion)
• Each strip of web or flange that contains perforations is replaced by a solid strip of reduced thickness
• Global and local buckling, $k_g$ = 0.6:

$$t_g = k_g\, t \left(\frac{L_{np}}{L}\right)$$

• Distortional buckling, $k_d$ = 0.8:

$$t_d = k_d\, t \left(\frac{L_{np}}{L}\right)^{1/3}$$

• $L_{np}$ is the solid length between holes, $L$ the pitch

Source: ANSI MH16.1-2023, Figure 8.2-1, example of perforation strips in a column

• One idealized section feeds the frame model (§7.2.4), the global buckling check, and the finite strip model for $P_{crd}$ and $M_{crd}$

• $P_n = \min(P_{nlg},\, P_{nld})$
• $P_{crd}$: elastic distortional buckling "in accordance with ANSI/AISI S100" on the $t_d$ section
• Hot-rolled and closed sections: $P_{nlg}$ only

Both forms with $F_n/F_y = P_{ne}/P_y$ on the same area basis. 2012 used $A_{net\,min}$; 2021 uses $A_{netg}$ from the strip section.

• Local buckling is not a DSM check: there is no $P_{cr\ell}$. The perforation and local slenderness effects both come from the stub-column test (§13.2.3, $Q = P_{test}/(F_y A_{netg})$ ≤ 1, interpolated between $t_{min}$ and $t_{max}$)
• The 2021 exponent $Q/(1-Q)$ makes the local penalty fade as the column becomes globally slender, the same trend as the DSM local–global curve
• For a given $Q$, the new form is less severe at intermediate $\lambda_c$

• New in 2021: an explicit distortional check with an elastic buckling load, $P_{crd}$ or $M_{crd}$, the DSM ingredient
• Same 0.25 / 0.6 coefficients and 0.561 limit as the AISI S100 distortional curve, but anchored at $P_{ne}$ instead of $P_y$
• That captures distortional–global interaction, which tests and FE studies on perforated uprights showed (Casafont, Pastor, Roure, Bonada, Peköz)
• At $\lambda_c$ = 0 the two agree; at intermediate slenderness MH16.1 is lower, and both reach $P_{ne}$ for slender columns

Example with $P_{crd} = 0.6\,P_y$

Source: ANSI MH16.1-2023 Commentary, Figure C8.2-5. CUFSM: local minimum at 80 mm (load factor 14.92), distortional at 700 mm (10.63)

• Commentary C8: the "most suitable design approach would be to use finite strip analysis combined with the expressions given"
• Workflow
  • build the round-corner section with $t_d$ in the perforated strips
  • run CUFSM (or any S100 Appendix 2 analysis)
  • read $P_{crd}$ at the distortional minimum
  • the same model with $t_g$ gives the global section properties
• Basis: tests, shell FE and FSM studies on perforated uprights (Casafont et al., ASCE J. Struct. Eng. 2013, "Design of steel storage rack columns via the Direct Strength Method")

• §7.2: second-order analysis with notional loads replaces the effective length method
  • $N_i$ = 0.004 $\alpha Y_i$ with the stiffness reduction $\tau_b$, or 0.005 $\alpha Y_i$ without; minimum 0.002 $\alpha Y_i$
  • $B_2 = 1 / \left(1 - \alpha P_\Delta / (R_m H L)\right)$
  • semi-rigid beam-to-column connections modeled with springs
• Down-aisle $K_x$ = 1.0 (§10.2.1)
• Member and frame models use the same hole-reduced section properties

• Connectors from cyclic tests (§13.5): design moment 0.75 $M_{max}$; stiffness is the secant at 0.8 $M_{conn,d}$
• New tests: base fixity (§13.6) and frame bracing (§13.7); portal and upright frame tests removed
• Seismic aligned to ASCE 7-16; redundancy factor for multiple rows; overstrength design of base plates and anchors (§11.3)
• ISO-style reorganization of the whole document